The Burden of Weight on Joint Replacement: A 1.6 Million-Patient Analysis of BMI and Hip Arthroplasty Outcomes

Abstract

Background: THA is a gold-standard intervention for end-stage hip osteoarthritis, historically performed in older adults. However, the growing global obesity epidemic is reshaping this landscape. Emerging evidence suggests that elevated body mass index (BMI) may not only worsen perioperative outcomes but also accelerate the need for surgery at a younger age. Understanding how BMI influences both the timing and safety of THA is crucial to optimizing care in this evolving patient population. Methods: We conducted a retrospective analysis of 1,626,965 elective THA hospitalizations from the Nationwide Inpatient Sample. Patients were stratified by BMI into three categories: <29.9, 30–34.9, and ≥35. Fracture- and oncology-related cases were excluded. ICD-10 codes identified comorbidities and complications. Primary outcomes included age at surgery, in-hospital mortality, length of stay (LOS), complications, and hospitalization costs. Statistical analysis used Pearson correlation, linear regression, chi-square tests, and t-tests via SPSS version 26.0.0.0. Results: Higher BMI was significantly associated with younger age at THA (r = −0.187, p < 0.001). Each 5-unit BMI increase corresponded to a ~2-year decrease in age at surgery. Obese patients had higher rates of hypertension, diabetes, dyslipidemia, and sleep apnea. Complications including blood loss anemia, acute kidney injury, venous thromboembolism, and postoperative infections were more common in higher BMI groups. LOS increased with BMI, though total hospital charges showed minimal clinical variation. Conclusions: Obesity is a key driver of earlier THA and elevated perioperative risk. These findings underscore the need for BMI-tailored surgical planning and risk stratification. As younger, high-BMI patients increasingly undergo THA, future strategies must focus on preoperative optimization, complication prevention, and long-term implant durability.

1. Introduction

Osteoarthritis is a chronic, degenerative joint disorder and a leading contributor to pain and mobility limitations in over 27 million Americans [1]. Total hip arthroplasty (THA) remains the definitive surgical treatment for advanced hip osteoarthritis—delivering sustained improvements in pain, mobility, and quality of life [2,3,4]. However, the obesity epidemic has significantly reshaped the patient profile for THA. Obesity not only accelerates the development of osteoarthritis [5,6,7] but also compounds surgical complexity, elevating both perioperative risks and long-term revision rates [8,9,10,11,12,13,14,15,16].

BMI, a standardized weight-for-height metric [17], has emerged as a critical variable influencing not just outcomes but also the timing of surgical intervention. Historically reserved for older patients [18], THA is now increasingly performed at younger ages—likely driven by obesity-related joint degeneration [15,19]. This shift introduces unique challenges: younger, high-BMI patients face increased implant stress, longer postoperative life expectancy, and higher functional demands—factors linked to early prosthesis failure and the need for complex revisions [18,20].

In this study, we leveraged the Nationwide Inpatient Sample (NIS) to evaluate the relationship between BMI and age at THA, along with associated perioperative outcomes, complications, length of hospital stay, and healthcare costs. By analyzing this large-scale, nationally representative dataset, we aim to provide actionable insights for improving patient-specific care strategies and informing health policy as the THA demographic continues to shift toward younger and more obese populations.

Research Questions

This study explores the primary determinants of THA across various BMI categories, with an emphasis on the influence of demographic and clinical characteristics on surgical outcomes. The research examines the prevalence and impact of comorbidities on postoperative complications while analyzing differences in hospital length of stay, healthcare costs, and mortality rates among BMI groups. Furthermore, it identifies common postoperative complications and investigates the age-specific effects of BMI on THA outcomes.

2. Methods

2.1. Data Source

The data were drawn from the Nationwide Inpatient Sample (NIS), a part of the Healthcare Cost and Utilization Project (HCUP). The NIS provides a representative 20% stratified sample of all inpatient discharges from US hospitals, amounting to approximately seven million unweighted hospital stays annually. This study analyzed data from 1 January 2016 to 31 December 2019.

2.2. Study Population and Variables

This study utilized a dataset comprising 1,626,965 cases of total hip arthroplasty (THA). Cases involving fractures and oncology were removed to ensure a focused analysis of elective THA. Fracture-related THA cases were excluded to isolate elective, degenerative indications, as inclusion of trauma cases would have introduced heterogeneity in urgency, comorbidity profile, and surgical complexity. Obesity and BMI classifications were identified using specific ICD-10 codes. The study population was categorized into the following BMI groups: BMI < 29.9, 29.9 < BMI < 34.9 (Class I obesity), 34.9 < BMI < 39.9 (Class II obesity), and BMI ≥ 39.9 (Class III obesity).

2.3. Correlation Between BMI and Age of THA

The relationship between BMI and the age of THA was analyzed using Pearson correlation and linear regression. This analysis aimed to determine whether there was a significant correlation between BMI and the age at which patients underwent THA. The Pearson correlation coefficient (r) and the significance level (p-value) were calculated to assess the strength and significance of the relationship. Linear regression analysis was performed to quantify the effect of BMI on the age of THA and to develop a regression equation.

2.4. Patient Identification and Comorbidities

Comorbidities were identified through a comprehensive review of patient-specific ICD-10 codes. Osteoporosis prevalence was based solely on ICD-10 diagnostic coding, as bone mineral density information is not available in NIS. Cases reporting hospital costs of $0 were excluded from the analysis. The following comorbidities were identified: hypertension, dyslipidemia, chronic anemia, osteoporosis, alcohol abuse, type 2 diabetes, renal disease, chronic heart failure (CHF), chronic lung disease, and use of anticoagulants.

2.5. Outcome Measures

The primary outcome measures in this study included in-hospital mortality, length of stay (LOS), complications, and total hospitalization costs. The NIS database reports total hospital charges but does not differentiate between implant, surgical, or facility costs; therefore, cost breakdown by category was not possible. Clinical outcomes were analyzed using established methodologies.

2.6. Statistical Analysis

Statistical analyses were conducted using SPSS version 26.0.0.0. Frequencies and proportions of different demographic, clinical, and hospital-related variables were calculated and weighted to reflect national estimates using discharge sample weights provided by the NIS. Comparisons were made using Pearson’s chi-square test for categorical variables and independent sample t-tests for continuous variables. A significance level of p < 0.05 was applied. Analytical studies were conducted using SPSS version 26.0.0.0 and Microsoft Excel to visualize annual cases, discern trends, and derive key statistical insights. The regression model evaluating BMI versus age at THA was unadjusted and intended to demonstrate the direct association between these continuous variables. Multivariable adjustment for comorbidities or socioeconomic factors was beyond the scope of this descriptive analysis. Given the large sample size, many comparisons achieved statistical significance with minimal absolute differences; therefore, effect size interpretation and clinical relevance were emphasized in the Section 4. Due to the difference in average age in BMI groups and the connection between BMI and age of THA, shown earlier in the article, we added an extra analysis to subgroups based on age to better understand these trends. Traditional statistical tests were applied to maintain consistency with prior national arthroplasty studies and to ensure clear, reproducible clinical interpretation. Advanced predictive modeling techniques were beyond the scope of this descriptive population analysis.

2.7. Ethical Aspects

The study was conducted under exempt status granted by the institutional review board, and the requirement for informed consent was waived due to the de-identified nature of the NIS dataset.

3. Results

3.1. BMI Classification and Distribution

The study population was categorized into four BMI groups: BMI < 29.9, 29.9 < BMI < 34.9 (Class I obesity), 34.9 < BMI < 39.9 (Class II obesity), and 39.9 < BMI (Class III obesity). Comprehensive tabular data and a regression scatter plot (Figure 1) were used to illustrate the main findings for optimal clarity and conciseness. The total number of patients in each BMI group is as follows: BMI < 29.9: 1,244,050; 29.9 < BMI < 34.9 (Class I obesity): 182,780; 34.9 < BMI < 39.9 (Class II obesity): 200,010; and 39.9 < BMI (Class III obesity): 125. Because the Class III obesity subgroup included fewer than 0.01% of total THA cases (n = 125), it was merged with Class II obesity to ensure sufficient statistical power and stable regression estimates. The two groups were rechecked and found to demonstrate parallel trends.

3.2. Correlation Between BMI and Age of THA

The relationship between BMI and the age at which patients undergo THA was assessed using Pearson correlation and linear regression analysis. A significant negative correlation was identified between BMI and age at THA (r = −0.187, p < 0.001). Additionally, linear regression analysis demonstrated that BMI is a significant predictor of the age at THA (β = −0.405, p < 0.001). The regression equation is as follows:

Age of THA = 77.989 − 0.405 × BMI

From this regression model, we can infer that for every 5-point increase in BMI, the age at which patients undergo THA decreases by approximately 2 years.

The scatter plot (Figure 1) visually illustrates the inverse relationship between BMI and the age of THA.

3.3. Baseline Characteristics of Patients by BMI Classification

The baseline characteristics of patients, categorized by BMI groups, are detailed in

Table 1. Baseline Characteristics of Patients by BMI Classification.

Parameter	BMI < 29.9	29.9 < BMI < 34.9 Class I Obesity	34.9 < BMI Class II Obesity + Class III Obesity	Significance
Total Surgeries (%)	76.50%	11.20%	12.30%	-
Average Age (y)	66.14	64.86	62.04	p < 0.0001
Female (%)	55.5%	51.8%	57.2%	p < 0.0001
Payer—Medicare (%)	56.5%	52.2%	45.4%	p < 0.0001
Payer—Medicaid (%)	4.8%	5.0%	6.5%
Payer—Private (%)	35.6%	39.8%	44.8%
Payer—Other (including self-pay) (%)	3.1%	3.0%	3.3%
Median Household Income 0–25% (%)	19.0%	18.6%	21.6%	p < 0.0001
Median Household Income 26th–50th% (%)	24.9%	23.9%	27.4%
Median Household Income 51st–75th% (%)	27.2%	27.8%	27.9%
Median Household Income 76th–100th% (%)	28.9%	29.7%	23.1%

. The majority of surgeries were performed on patients with a BMI of less than 29.9, comprising 76.5% of the total surgeries. The average age at the time of surgery decreased as BMI increased, from 66.14 years for patients with a BMI of less than 29.9 to 62.04 years for those with Class II and Class III obesity (p < 0.0001). Furthermore, a higher proportion of females was observed in the Class II and Class III obesity groups, accounting for 57.2%.

A majority of patients with a BMI less than 29.9 and Class I obesity were Medicare beneficiaries, whereas a higher percentage of patients in the Class II and Class III obesity groups were covered by private insurance (44.8%) (p < 0.0001).

3.4. Etiologies of Total Hip Arthroplasty by BMI Classification

The etiologies leading to THA are summarized in

Table 2. Etiologies of Total Hip Arthroplasty by BMI Classification.

Etiologies	BMI < 29.9	29.9 < BMI < 34.9	34.9 < BMI	Significance
Primary osteoarthritis	91.42%	93.09%	93.63%	p < 0.0001
Revision	4.10%	3.51%	3.65%
Osteonecrosis	4.11%	3.08%	2.41%
Rheumatoid arthritis	0.16%	0.13%	0.10%
Post-traumatic arthritis	0.11%	0.09%	0.09%
Legg-Calvé-Perthes	0.05%	0.04%	0.05%
Leg Deformity	0.04%	0.05%	0.07%

. Primary osteoarthritis is the predominant reason for THA across all BMI categories, with its prevalence slightly increasing with higher BMI (91.42% for BMI < 29.9, 93.09% for BMI 29.9–34.9, and 93.63% for BMI ≥ 34.9, p < 0.0001).

Revision surgeries and osteonecrosis show a decrease in prevalence with increasing BMI. Rheumatoid arthritis, post-traumatic arthritis, Legg–Calvé–Perthes disease, and leg deformities are relatively rare etiologies with slight variations across BMI categories.

3.5. Comorbidities by BMI Classification

The prevalence of various comorbidities across different BMI categories is summarized in

Table 3. Comorbidities by BMI Classification.

	BMI < 29.9	29.9 < BMI < 34.9	34.9 < BMI	Significance
Hypertension Diagnosis	49.5%	57.7%	63.4%	p < 0.0001
Dyslipidemia Diagnosis	40.3%	51.1%	47.3%	p < 0.0001
Sleep Apnea Diagnosis	6.7%	15.3%	26.1%	p < 0.0001
Chronic Anemia	5.7%	5.8%	5.9%	p < 0.0001
Alcohol Abuse	1.5%	1.6%	1.1%	p < 0.0001
Osteoporosis	5.00%	3.80%	2.30%	p < 0.0001
Mental Disorders	29.2%	33.0%	34.5%	p < 0.0001
Type 2 Diabetes	12.3%	20.4%	26.7%	p < 0.0001
Renal Disease	5.8%	8.4%	8.4%	p < 0.0001
CHF	1.1%	1.4%	1.7%	p < 0.0001
Chronic Lung Disease	6.5%	6.9%	7.5%	p < 0.0001
Use of anticoagulants	5.2%	6.4%	7.2%	p < 0.0001

Abbreviations: CHF—Congestive Heart Failure.

. Hypertension, dyslipidemia, and sleep apnea diagnoses increase with higher BMI. Other notable trends include an increase in type 2 diabetes and the use of anticoagulants with higher BMI, while the prevalence of osteoporosis decreases as BMI increases. Mental disorders and chronic lung disease also show higher prevalence in patients with higher BMI.

3.6. Complications and Outcomes by BMI Classification

The prevalence of various complications and outcomes across different BMI categories is summarized in

Table 4. Postoperative Complications by BMI Classification.

	BMI < 29.9	29.9 < BMI < 34.9	34.9 < BMI	Significance
Blood Loss Anemia	17.49%	19.09%	19.15%	p < 0.0001
Blood Transfusion	3.00%	2.46%	2.53%	p < 0.0001
Acute Kidney Injury	1.50%	2.00%	3.10%	p < 0.0001
Intraoperative Fracture	0.77%	0.72%	0.89%	p < 0.0001
Venous Thromboembolism	0.16%	0.20%	0.24%	p < 0.0001
Hip Dislocation	0.15%	0.15%	0.13%	p = 0.044
Acute Coronary Artery Disease	0.09%	0.08%	0.06%	p = 0.001
Pneumonia	0.13%	0.14%	0.15%	p = 0.031
Pulmonary Embolism	0.07%	0.09%	0.11%	p < 0.0001
Heart Failure	0.08%	0.14%	0.13%	p < 0.0001
Pulmonary Edema	0.04%	0.04%	0.07%	p < 0.0001
Infection	0.03%	0.03%	0.04%	p = 0.013

. Blood loss anemia, acute kidney injury, and venous thromboembolism increase with higher BMI. Blood transfusion rates slightly decrease with higher BMI. The incidence of intraoperative fractures and infections is also slightly higher in the highest BMI category.

3.7. Mortality, Length of Stay, and Charges by BMI Classification

Key outcomes related to hospitalization are presented in

Table 5. Mortality, Length of Stay, and Charges by BMI Classification.

	BMI < 29.9	29.9 < BMI < 34.9	34.9 < BMI	Significance
Died during hospitalization	0.04%	0.03%	0.03%	0.087
Length of stay mean in days	2.00	2.01	2.21	p < 0.0001
Total charges mean in $	63,490	63,920	63,819	p < 0.0001

. The mortality rate during hospitalization is similar across all BMI groups. However, the length of stay increases with higher BMI, averaging 2.00 days for BMI < 29.9, 2.01 days for BMI 29.9–34.9, and 2.21 days for BMI ≥ 34.9 (p < 0.0001). Total charges significantly differ statistically, with mean charges of $63,490 for BMI < 29.9, $63,920 for BMI 29.9–34.9, and $63,819 for BMI ≥ 34.9 (p < 0.0001). However, hospital charges represent billed amounts rather than actual costs, which vary between institutions; thus, these statistical differences likely reflect administrative rather than true economic variation.

3.8. Age-Specific Complications and Outcomes by BMI Classification

Table 6. Age-specific Complications and Outcomes by BMI Classification.

	BMI < 29.9	29.9 < BMI < 34.9	34.9 < BMI	Significance
Blood Loss Anemia (age under 60)	15.2%	17.0%	17.5%	p < 0.0001
Blood Loss Anemia (age 60–70)	16.1%	18.1%	19.1%	p < 0.0001
Blood Loss Anemia (age over 70)	20.6%	22.6%	22.8%	p < 0.0001
Blood Transfusion (age under 60)	2.4%	2.1%	2.3%	p < 0.0001
Blood Transfusion (age 60–70)	2.4%	2.0%	2.3%	p < 0.0001
Blood Transfusion (age over 70)	4.0%	3.4%	3.4%	p < 0.0001
Acute Kidney Injury (age under 60)	0.8%	1.1%	1.9%	p < 0.0001
Acute Kidney Injury (age 60–70)	1.1%	1.7%	3.3%	p < 0.0001
Acute Kidney Injury (age over 70)	2.5%	3.4%	5.4%	p < 0.0001
Intraoperative Fracture (age under 60)	0.7%	0.6%	0.9%	p < 0.0001
Intraoperative Fracture (age 60–70)	0.7%	0.6%	0.8%	p < 0.0001
Intraoperative Fracture (age over 70)	0.9%	0.9%	1.1%	p = 0.001
Length of stay mean in days (age under 60)	1.85	1.86	2.08	p < 0.0001
Length of stay mean in days (age 60–70)	1.86	1.91	2.2	p < 0.0001
Length of stay mean in days (age over 70)	2.25	2.3	2.52	p < 0.0001
Total charges mean in $ (age under 60)	63,624	64,067	63,741	p < 0.0001
Total charges mean in $ (age 60–70)	62,658	62,591	63,423	p = 0.052
Total charges mean in $ (age over 70)	64,183	65,425	64,724	p < 0.0001

presents a comprehensive analysis of age-specific complications and outcomes across BMI classifications. Due to the difference in average age in BMI groups and the connection between BMI and age of THA, shown earlier in the article, an additional analysis was conducted on age-based subgroups to further elucidate these trends. The analysis reveals that complications such as blood loss anemia, acute kidney injury, and intraoperative fractures are significantly higher in older patients with higher BMI. Furthermore, the length of stay and total charges increase with both age and BMI, highlighting the compounded risk factors that age and obesity contribute to surgical outcomes.

4. Discussion

4.1. Main Findings

Our main finding indicates that obese patients were significantly younger, supporting the association between obesity and early-onset osteoarthritis. Additionally, a disproportionately high rate of female patients (57.2%) with BMI ≥ 34.9 was noted, consistent with prior observations of osteoarthritis prevalence in obese women. Furthermore, obesity was associated with an increased prevalence of hypertension, dyslipidemia, sleep apnea, type 2 diabetes, and chronic lung disease. The incidence of complications, including blood loss anemia, acute kidney injury, venous thromboembolism, intraoperative fractures, and postoperative infections, was also significantly higher in obese patients. Interestingly, osteoporosis prevalence and the need for blood transfusions decreased with increasing BMI. This inverse association may be related to lower relative blood loss in obese patients or to institution-specific restrictive transfusion thresholds not captured within NIS data; however, this remains speculative and warrants future confirmation. While hospitalization length increased, total hospital charges remained relatively unaffected.

4.2. Time to THA

Despite the well-established pathophysiological effects of obesity on osteoarthritis, its influence on the age at which THA is performed requires further investigation.

Singh et al. [21] analyzed data from the Mayo Clinic, including 6168 patients, and showed that between 1993 and 2005, the mean age at THA decreased by 0.7 years (p < 0.002) while BMI increased by 1.6 kg/m² (p < 0.001).

Similarly, Haynes et al. [15] performed a systematic review of 17 studies encompassing 13,722 THA patients and found a consistent trend of younger surgical age among obese individuals.

Johnson et al. [22,23] analyzed the National Inpatient Sample database and identified 688,371 THA and 1,556,651 TKA cases over sixteen years, demonstrating a significant decrease in mean age for THA (from 66.7 to 65.9 years, p < 0.001) alongside a rise in the proportion of obese patients undergoing these procedures (from 7.0% to 22.7%, p < 0.001).

The present study corroborates these findings, revealing that patients with higher BMI were significantly younger at the time of surgery. The mean age at surgery decreased from 66.14 years in patients with BMI < 29.9 to 62.04 years in those with Class II and Class III obesity (p < 0.0001). This suggests that obesity accelerates the need for surgical intervention due to its exacerbating effects on hip osteoarthritis, leading to more rapid joint degeneration.

This trend is concerning, as younger age at initial THA increases the likelihood of revision surgery, which is technically challenging, more expensive, and associated with inferior outcomes compared to primary THA [24,25,26].

4.3. Complications

Obesity is widely believed to increase perioperative and postoperative complications, though literature findings remain mixed.

Moran et al. [27] reviewed 800 THA cases followed for at least 18 months and found no correlation between BMI and complications, attributing this to regression analysis that adjusted for confounders but noting limited power for rare events.

Contrary to those findings, our study demonstrated that obese patients had significantly higher complication rates than their non-obese counterparts. Specifically, blood loss anemia, acute kidney injury, venous thromboembolism, intraoperative fractures, and postoperative infections were all significantly associated with higher BMI, aligning with prior reports [8,9,10,11,12,13,14,15,16]. Interestingly, the need for blood transfusions decreased with increasing BMI, suggesting a complex interplay between obesity and surgical bleeding risk. This contrasts with Friedman et al. [28], who reported no association between morbid obesity and increased bleeding or VTE risk, underscoring variability in perioperative practices.

Given the low absolute rates of major complications, large-scale studies such as ours are essential for detecting significant differences [29]. The elevated complication rate in obese patients may be attributed to the technical challenges of THA in this population. The increased mass of adipose and muscle tissue necessitates prolonged surgical exposure and prosthesis insertion time. Additionally, higher BMI contributes to increased mechanical loading on hip prostheses, which could accelerate implant loosening. However, this effect may be counterbalanced by the relatively lower activity levels of morbidly obese patients, which could reduce prosthetic wear [30,31,32].

Moreover, obesity-related metabolic changes, including denutrition and chronic low-grade inflammation, may predispose patients to complications [22,33,34,35]. Further research is required to elucidate these metabolic alterations and their clinical implications in THA.

4.4. Hospitalization Factors

The impact of obesity on LOS and overall costs remains debated. Given the increasing prevalence of obesity and joint arthroplasties, understanding their economic implications is essential. Some studies found no association between obesity and LOS [36,37,38], while others reported increased resource utilization in obese patients [39,40].

Wayne A. Wilkie et al. analyzed 453,250 National Inpatient Sample cases of THA performed between 2009 and 2016 in patients with BMI > 25 kg/m². They found that morbidly obese patients (BMI > 40.1 kg/m²) had significantly longer LOS compared to non-morbidly obese (BMI 30.0–40.0 kg/m²) and overweight (BMI 25.0–29.9 kg/m²) patients (p < 0.001) [41]. These findings are consistent with the present study, suggesting that obesity increases the complexity of perioperative care and recovery, necessitating extended hospitalization and additional monitoring.

Despite the prolonged LOS, this study found that total hospital charges did not significantly vary across BMI categories. The NIS provides total hospitalization charges but not itemized billing data such as implant, surgical, or facility costs. Implant expenses are known to represent a major portion of arthroplasty expenditures in the U.S. and are under ongoing national cost-reduction initiatives. This contradicts the findings of Senthil Sambandam et al., who, in a retrospective National Inpatient Sample (NIS) analysis of 1646 THA cases, reported significantly higher costs in the super-obese group (BMI ≥ 50 kg/m²; $79,784.64) compared to the non-super-obese group ($66,821.75) [40]. The discrepancy may be due to differing BMI classification methods.

Given the broader dataset analyzed in this study, its findings suggest that while obesity leads to increased healthcare resource utilization, the additional financial burden per patient may be less significant than previously thought.

4.5. Limitations and Strengths

This study benefits from a large, nationally representative dataset, providing robust insights into the relationship between BMI and THA outcomes. However, several limitations should be noted. Because NIS captures only the index hospitalization, long-term outcomes such as revisions, readmissions, or functional scores could not be evaluated. Osteoporosis and other comorbidities were identified using diagnostic codes, which may introduce misclassification. Although female predominance was observed in higher BMI groups, sex-specific subgroup analyses were not performed. Finally, results should be interpreted with consideration of the dataset’s administrative nature and limited clinical granularity. Because NIS lacks longitudinal identifiers, revision surgeries performed after discharge could not be analyzed, limiting assessment of long-term associations between BMI and implant survival.

5. Conclusions

Understanding the impact of BMI on the timing, complications, and outcomes of total hip arthroplasty is crucial for optimizing patient care.

• Higher BMI is significantly associated with younger age at THA.
• Obesity increases perioperative complications, including anemia, acute kidney injury, and venous thromboembolism.
• Hospital stay length rises with BMI, while total hospital charges remain relatively stable.
• These findings emphasize the importance of BMI-tailored perioperative optimization and long-term risk assessment.